Electrical Grid Stability Simulated Data

Overview

The Electrical Grid Stability Simulated Data set contains simulated measurements from a 4-node star system (with the electricity producer at the center) implementing the Decentral Smart Grid Control concept.

Data Collection

• 10,000 instances generated using simulations based on differential equations for the system's dynamic behavior.
• The analysis follows methodologies similar to those in power grid stability research, with fixed parameters for averaging time (2 s), coupling strength (8 s⁻²), and damping (0.1 s⁻¹).
• Each instance represents a snapshot of the grid state and includes a stability label (stable/unstable).

Variables

• tau[1-4]: Reaction time of each participant (real from the range [0.5,10] s). tau1 is the value for the electricity producer.
• p[1-4]: Nominal power consumed (negative) or produced (positive). For consumers, from the range [-0.5,-2] s⁻²; p1 = abs(p2 + p3 + p4).
• g[1-4]: Coefficient (gamma) proportional to price elasticity (real from the range [0.05,1] s⁻¹). g1 is for the producer.
• stab: The maximal real part of the characteristic equation root (if positive, the system is linearly unstable).
• stabf: The stability label of the system (categorical: stable/unstable).

Use Cases

• Classification and regression tasks for predicting smart grid stability.
• Evaluating machine learning models on power system control and optimization problems.
• Research on decentralized control strategies and grid resilience.